Submitted by Frank Miller on December 20, 2022 - 15:02
This course focuses on computational methods for optimisation, simulation and integration needed in statistics. The optimisation part discusses gradient based, stochastic gradient based, and gradient free methods. Further, constrained optimisation will be a course topic. We will discuss techniques to simulate efficiently for solving statistical problems. The course will start on March 16; course homepage (with full schedule): http://www.adoptdesign.de/frankmillereu/adcompstat2023.html.
At the Division of Statistics and Machine Learning, Department of Computer and Information Science, Linköping University, 21st-24th March 2023 we will be hosting a school concerning stochastic differential equations and the YUIMA R package (Simulation and Inference for SDEs and Other Stochastic Processes, https://cran.r-project.org/web/packages/yuima/index.html ) . The lectures will be given by members of the YUIMA team. Below is a nearly final program of the school. The dates are fixed.
Submitted by johantykesson on September 23, 2022 - 01:39
Course description
The course will consist of two or three cases, i.e., scientific questions that we seek to answer. Scientific background material, in the form of links, summary texts, and lectures, will be provided to aid the analysis. Technical lectures (detailed or overviews) are available upon request. The class will be divided into groups, each of which do their own analysis. The intent is to have each group contain a broad spectrum of expertise.
This will be a PhD course in probability theory taught by Zangin Zeebari. The course will be online. For further question and registration contact Kristofer Månsson (kristofer.mansson@ju.se). The syllabus contains information about content and litterature etc.
Textbook The course will mainly follow A. W. van der Vaart (1998) Asymptotic Statistics. Cambridge University Press, Cambridge, UK.
Structure Lectures and exercises Assessment Hand-in exercises.
Pre-requisites Statistical Inference I (7.5 hp), PhD level, or equivalent.
Course plan Selected chapters from van der Vaart (1998) will be covered during the course. The course will be held during June (weeks 23, 24, 25) and August (weeks 33 and 34).
Submitted by jolanta.pielasz... on February 6, 2021 - 09:58
Course plan
The course is about optimal experimental design – planning of experiments. The basic idea of design optimization (best estimation of unknown model parameters, information or moment matrix, etc.) and commonly used design criteria in linear models are the main parts of the course. Besides optimal designs in classical linear models, optimal designs for estimation and prediction of fixed and random effects in particular mixed models will be discussed.
Submitted by Frank Miller on January 8, 2021 - 12:55
Based on the topics discussed in first part of the course, we continue with deepening the theoretical basis for stochastic optimisation algorithms. Specifically, we discuss theory around Stochastic Gradient Ascent (including momentum and adaptive step sizes), Simulated Annealing, and Particle Swarm Optimisation. Theoretical results on convergence and speed will be discussed.
Submitted by xavierdeluna on October 5, 2020 - 13:20
This course will be run on-line via zoom, except for the final exam which is onsite (but local examination might be arranged for PhD students at other universities).
Submitted by larsronn on September 18, 2020 - 08:06
Upon completion of the course, the doctoral student will be able to: explain the basic theoretical foundation and practical use of Structural Equation Models (SEM), apply SEM on real problems, interpret and present the results, estimate structural equation models with maximum likelihood and least squares and be able to evaluate the results.
